PROCEDURE REPRODUCTIVE SUCCESS BY INDIRECT EVIDENCE OF APPARENT OCCUPANCY (CHOWIET, ST. LAZARIA)

Wildlife Inventory Plan Alaska Maritime National Wildlife Refuge Protocol #8 Version 1.4 Parameter: Productivity, phenology, and chick growth Species: Rhinoceros auklet PURPOSE To estimate annual productivity and phenology of burrow-nesting rhinoceros auklets and track differences among years. Seabird reproductive parameters can serve as indicators of change in the marine ecosystem (Cairns 1987, Montevecchi 1993). In addition, at some sites, to examine patterns in growth rates of rhinoceros auklet chicks as a potential indicator of foraging conditions for adults. Chick growth rate in many seabird species has been associated with spatial and temporal variability in food availability (e.g., Suryan et al. 2002). BREEDING BIOLOGY Rhinoceros auklets (Cerorhinca monocerata) are colonial, piscivorous seabirds that breed on remote islands in Alaska, Japan, Russia, and south to California. The vast majority of birds breed in British Columbia; the Gulf of Alaska represents the northern extent of their distribution. Rhinoceros auklets are closely related to Fratercula puffins but are nocturnal at Alaskan breeding colonies, appearing at dusk and leaving again before dawn. All colony social activity takes place at night (Gaston and Dechesne 1996). Rhinoceros auklets nest primarily in soil burrows, where they lay a single egg. The incubation period lasts about 45 days, after which chicks are brooded for about four days before adults leave their chick unattended in the burrow while foraging. Chicks fledge at about 50 days of age. Both parents share incubation and chick provisioning duties. (Leschner 1976, Wilson 1977, Summers and Drent 1979). Timing of breeding varies across sites, with later phenology at higher latitudes. At colonies in Alaska, rhinoceros auklets generally lay eggs in May and June, hatch chicks from mid-june through July, and fledge chicks in August and September (Gaston and Dechesne 1996, Drummond and Slater 2012). PROCEDURE REPRODUCTIVE SUCCESS BY INDIRECT EVIDENCE OF APPARENT OCCUPANCY (CHOWIET, ST. LAZARIA) Data collection. When burrows are too deep to access nest contents, indirect evidence of burrow apparent occupancy at different times in the breeding season is used as an index of rhinoceros auklet reproductive performance. We assume that apparent occupancy during early and late season checks represent the presence of eggs and large chicks/potential fledglings, respectively. From these data, a residency index can be calculated (representing a surrogate of reproductive success). These data are usually collected simultaneously with population surveys for burrow density and apparent occupancy (see Burrow-nester Population Protocol and island-specific details). Burrows are checked twice during the season, timed to occur around hatching and during late chick-rearing (these are often done in conjunction with population surveys, which have an additional third check earlier in the season at some sites, so don t be confused if you check rhinoceros auklet plots three times at your island data for reproductive success will just come from the later two). On each check, observers locate all medium (9.5-14.5 cm) and large (>14.5 cm) burrows within specified plots and assess each for any evidence of apparent occupancy. Burrow size is categorized as small, medium, or large (small: <9.5 cm, medium: 9.5-14.5 cm, large: >14.5 cm) and measured using a burrow board or burrow measuring device (Figure 1). Measure each entrance horizontally at the widest spot: if the 8-1

smallest end of the board does not slide in to the entrance, it is assigned to the small category; if the small end will slide in, it is assigned to the medium category; if the largest section slides in, it is assigned to the large category. By definition, a burrow must be at least 30 cm deep to be counted. You can use your arm as a measuring device for burrow depth (starting at your finger tips, determine where 30 cm hits your arm, for an average-sized person elbow length will suffice). Assign a status of apparently occupied or unoccupied to each burrow on each visit, based on indirect (usually) evidence that the nest site is being used by a breeding pair. Evidence of apparent occupancy is defined as the presence of at least one of the following within the burrow: feathers guano fish deep inside the burrow egg shell fragments or membrane(s) egg(s) chick(s) fresh vegetation at least 20 cm inside the burrow entrance (ST. LAZARIA ONLY) clipped vegetation or roots inside the burrow entrance (ST. LAZARIA ONLY) Also record which evidence was observed at each burrow (this will help us understand and improve the rhinoceros auklet protocol). Feathers should be of the proper color and size for a rhinoceros auklet (e.g., if other burrow nesters are present at your site, you should not count feathers known to be from other species). Guano will often be in a pocket at the end or side of a tunnel (if you can reach that far) but also can be present within the burrow section accessible from the entrance. Fresh digging and fish at the entrance are not indicators of occupancy. Fish presence is tricky because they may have been dropped by a bird flying over, so be sure fish are deep enough in a burrow that they couldn t have simply fallen from above before using them as evidence of occupancy (use a reasonable person s judgment that fish were brought into the burrow and not haphazardly dropped near/in the burrow s entrance). At sites with multiple burrow nester species, egg shell fragments and membranes should be of proper size and color. Fresh vegetation should be pulled inside the entrance or deep in the burrow. At some sites, a runway of packed vegetation or dirt may extend from a burrow entrance but this alone is not enough to determine apparent occupancy (you may record runways in your data collection, but don t use them for anything). All burrows should be marked with individually-numbered flags or other identifiers near the entrance. It is helpful to choose a standard flag location relative to the entrance for all your nests (e.g., always putting flag on right side of burrows as you face uphill; see island specifics for details on flag locations) to avoid confusing nests. In high density areas, tunnels and nest chambers may be everywhere, so be sure that the flag shaft does not pierce into the tunnel or chamber of another nest. It is helpful to create a map indicating approximate location and number of each burrow within your plot to assist in relocating burrows (Figure 2). Data analysis. From data, tally the following for each plot: Medium/large burrows with apparent occupancy Early (b) number of medium/large ( 9.5 cm) burrows with evidence of apparent occupancy at early check (around hatching); surrogate of number of eggs Medium/large burrows with apparent occupancy Late (f) number of medium/large ( 9.5 cm) burrows with evidence of apparent occupancy at late-season check (during late chickrearing) that also had evidence of apparent occupancy at the early-season check (b); surrogate of number of large chicks/fledglings From the above values, calculate for each plot: Overall Residency Index (f/b) surrogate of overall reproductive success 8-2

This index represents a ratio of nests occupied at the end of the season compared to at the beginning. Therefore, for this calculation it is important to use only those nests that were occupied during the first visit. If a nest was not apparently occupied the first visit but it was the last visit, this nest should be excluded. For example: Early visit Late visit Use in Overall Residency Index calculations? Occupied Occupied Yes Occupied Not occupied Yes Not occupied Occupied No Not occupied Not occupied No If your residency index is greater than 1, you are doing something wrong! 8-3

PROCEDURE PHENOLOGY AND CHICK GROWTH (ST. LAZARIA) Data collection. Rhinoceros auklet phenology is monitored by manually checking individually numbered burrows at 5-7 day intervals throughout the breeding season. Hatch dates are calculated either from nest observations or from chick growth measurements. Because rhinoceros auklet burrows are usually very long, deep, and difficult to monitor, artificial nest boxes may be used supplementally at some sites. Historically, rhinoceros auklet reproductive success was also estimated using direct observations of nest statuses at some sites, but this has been discontinued. Begin examining burrows and nest boxes for very young chicks and/or brooding birds in early chick-rearing (depends on site, see island-specific details for more information). Aim for a sample of 30 active nests for phenology and chick growth. Ideally, use only burrows for which you can examine in its entirety (obviously, this is usually not a problem with nest boxes). At some sites, you may use and make additional excavations to observe nest contents and extract chicks. This may be done only in areas where the soil depth is sufficient to support the observer s weight and where vegetation is robust. Do not examine burrows in areas where there is extensive bare ground between shrubs and trees (e.g., much of Rhino Point on St. Lazaria). Mark all active burrows and nest boxes with individually-numbered flags near the entrance. Always place the flag at the right side of the entrance (as you face uphill) to avoid confusing nests. In high density areas, tunnels and nest chambers may be everywhere, so be sure that the flag shaft does not pierce into the tunnel or chamber of another nest. Be aware that vegetation will grow throughout the season, sometimes a meter or more, so it may be helpful to use additional markers, such as flagging tape tied around vegetation, to find burrows in areas of thick vegetation. Drawing a map of nest locations will also help locate nests during subsequent visits. If an incubated egg is found during the first status check, wait 7-10 days before checking the burrow again. Any non-brooded chicks can be measured immediately. Aim to measure chicks every 5-7 days until you obtain three wing chord measurements. Thereafter, take all measurements every other visit (i.e., 10 days apart) until chicks have lost most of their down and have adult feathers (e.g., having just wisps of down (especially on the rump or flanks), a "topknot" of down on the crown, and/or a "necklace" [an indentation ringing the base of the neck]). Then increase visits to every five days again to obtain fledging data. Rhinoceros auklet chicks can be difficult to extract, especially when they get larger and more mobile. It may take two people to work one burrow, blocking certain paths that the chick may run to escape your grasp. In addition, be careful if measuring chicks in adverse conditions. Protect chicks from wind, rain, and the observant eyes of predators by using an umbrella when needed. Wind will also affect the accuracy of the scale. On each visit, record the status of nest sites and all measurement data. For nest statuses, record the status of each nest each visit using the appropriate standardized code (see pages 8-8 to 8-10). Record only what you saw and be sure to use the standardized codes exactly as instructed (e.g., you must literally see the chick to record the presence of a chick; hearing it alone or seeing eggshells/membranes is not enough). If a rhinoceros auklet adult is encountered in a burrow, quietly and carefully back away to avoid further disturbance; check the burrow again in 5-7 days to minimize disturbance Rhinos are highly sensitive to disturbance prior to egg-laying and during incubation! There is sometimes a rudimentary nest composed of Angelica, false lily-of-the-valley, or fern leaves. Rhinoceros auklets do not exhibit reliable incubating or brooding postures (nor are we often able to actually view them), so if you see/feel an adult bird but cannot reach under it or see an egg or chick, you must record the nest status as BU (Bird Unknown). If you do see/feel a rhino-sized egg or a chick, there are a few cases when you will need to record additional modifiers that provide more detailed information about the nest status (see list of standardized code modifiers). For eggs, record if you see an old egg clearly from last year (Ely), a broken, crushed, or otherwise dead egg (Ed), an egg ejected from the nest (Ej), or an egg pipping just before hatch (Ep). Similarly, for chicks, record if you see a chick in the actual act of hatching ( Co ). Bear in mind that Ep and Co statuses would be extremely rare observations because it s likely that a parent would be present and susceptible to disturbance (so we wouldn t intentionally be reaching into a burrow at that point). 8-4

For measurements, refer to Figure 3: Mass: Weigh chicks (to the nearest gram (g)) in bird bags using Pesola spring scales. Be sure to weigh the empty bag after each chick weighing, because its weight can change due to moisture or chick feces. Bag weight is subtracted from the total to yield the chick s mass. Keep a free hand under the chick as it s being weighed to catch it in case the scale clip fails. Relaxed wing chord: Measure the right wing from the bend in the wing to longest tip (pin feather or longest primary, depending on chick age; ignore any downy tufts stuck to the end of wings or feathers) to the nearest 1 mm using a wing rule. With the right wing resting naturally against the chick s body, slide the wing ruler under the wing and press the vertical stop gently against the wrist joint. Measure this distance to the nearest mm. Only begin measuring wing chord once pin feathers erupt. Take care: pin feathers are very sensitive and will bleed if broken, which can lead to chick death; over-zealous handling can maim the developing wing. **Note: An additional wing measurement of flattened wing chord (measured as described above, but with pressing wing flat against wing rule) was historically recorded at St. Lazaria but this has been discontinued due to inconsistency in measurements. Diagonal tarsus: Hold chick's right leg so both the foot and tarsal joints are bent at right angles. Place the inner jaw of the calipers into the notch at the tarsal joint and slide the outer jaw until it contacts the end of the tarsus just where the foot bends. Measure this distance to the nearest 0.1 mm. Be sure the scale has been calibrated and zeroed prior to using it and recheck that it s zeroed before weighing each chick. Similarly ensure calipers and scales are zeroed before use. Data analysis. Calculations for phenology and chick growth parameters for rhinoceros auklets are not yet automated in an Access database so analysis needs to be conducted by field crews. Phenology: Hatch dates for rhinoceros auklet phenology can be calculated by nest statuses or by chick wing chord measurements, depending on which data are available. Nest status method: If data contain enough observations of eggs, chicks, and empty nests, dates for chick hatching and fledging can be calculated using the midpoint dates. The hatch date is the midpoint between the last time an egg was confirmed to be present and the first time a chick was confirmed to be present (if there was no midpoint [i.e., an even number of days between visits], we use the even Julian date closest to the midpoint; in leap years, be sure to use a leap year-specific Julian date calendar!). Not all nest sites are included in phenology calculations; we require a confirmed egg and then a confirmed chick less than or equal to 7 days apart for that site to be used, or a reasonable calculation of estimated hatch date based on chick measurements. Because of this, try to project when each nest is expected to hatch and focus extra attention on getting known statuses with short visit intervals around those times. Occasionally, we record data that give us more exact information on hatching and fledging dates. If you observe an event occurring (e.g., the actual hatching and fledging) during your visit, we use the day of the observation as the date the event occurred and not the midpoint between observations. Similarly, if a pipped egg is observed, we assume it will hatch the following day; if a wet chick is seen, we assume it hatched that day. For example: Julian Dates:165 172 178 182 190 198 212 218 223 Hatch date Fledge date Nest 1 BE BE BC BC C C C C N 175 220 Nest 2 BE BEp C C C C C N N 173 (pipped egg) 214 Nest 3 BE BE BCw C Cd N N N N 178 (wet chick) - Nest 4* BE BU BC C C C C N N Not used, >7d interval 214 * Hatch date may be calculated if enough measurements were taken in the linear phase of growth (as described below). 8-5

Chick wing chord method: If nest status data have a lot of unknown statuses (i.e., BU and U) but chick growth measurements are available (see chick growth below), hatch dates can be estimated using wing chord measurements (Leschner 1976). Chick age at measurement can be calculated as: Chick age = ln [(165 / WC)-1] / (-0.07) + 27.8 where ln is the natural log and WC is a wing chord measurement taken during the linear growth period (50-80 mm). Then calculate hatch date by subtracting the chick age from the date it was measured. If a chick has more than one wing chord measurement during the linear growth period, average the calculated hatch dates to derive a single estimate for each chick. Chick growth: Chick growth is summarized during the linear growth period only. In rhinoceros auklets, this corresponds roughly to 5-30 days old. After this age, chick size approaches an asymptote. Because you may not know exact hatch dates and thus accurate chick ages, it is helpful to graph data for each chick and visually inspect graphs for any leveling off towards the end of measurements. Truncate any data points that appear to be beyond linear growth. [Note - if you have only three measurements, it may be difficult to determine from the plot if the last measurement is outside the linear growth window; in these instances, mass is a particularly useful indicator, i.e., chicks are usually within the linear growth phase when <350 g.] Use the r 2 value of the linear regression line as a guide to the data s goodness of fit. There is no concrete rule for this value. Closely examine growth data in which the r 2 value is <0.80. If the poorness of fit is due to an early or late data point and the fit can be improved markedly by removing it (see a below), consider that point outside of the linear growth period and remove it. If the poorness of fit seems to be due simply to variability in measurements (see b below), consider keeping all data points. If the poorness of fit is because the chick just didn t seem to exhibit any linear growth at all (this can happen with dying chicks, for instance; see c below), consider removing the entire chick from the sample, as we are only interested in chicks for which we can calculate a rate of linear growth. (a) (b) (c) Once all data have been inspected and truncated to include only those measurements occurring during the linear growth period, calculate the growth rate for each chick by running a simple linear regression for each parameter (mass, wing chords, diagonal tarsus). The slope of each regression line represents the growth rate. Present the average annual growth rate for each species as the mean slope (and standard deviation) of all chicks. Include only chicks that have been measured at least three times during the linear growth phase. If a chick dies after three measurements, it can be included in the dataset. Literature Cited Cairns, D.K. 1987. Seabirds as indicators of marine food supplies. Biological Oceanography 5:261-271. Drummond, B.A. and L. Slater. 2012. Biological monitoring at St. Lazaria Island, Alaska in 2011. U.S. Fish and Wildlife Service Report, AMNWR 2012/02. Homer, Alaska. Gaston, A.J. and S.B.C. Dechesne. 1996. Rhinoceros auklet (Cerorhinca monocerata). In Birds of North America, No. 212 (A. Poole and F. Gill, eds.). The Birds of North America, Philadelphia. Leschner, L.L. 1976. The breeding biology of the rhinoceros auklet on Destruction Island. MS thesis, 8-6

Univ. Washington, Seattle. Montevecchi W.A. 1993 Birds as indicators of change in marine prey stocks. Pp 217-266 in Birds as Monitors of Environmental Change (R.W. Furness and D.J. Greenwood, Eds.) London: Chapman and Hall. Summers, K.R. and R.H. Drent. 1979. Breeding biology and twinning experiments of rhinoceros auklets on Cleland Island, British Columbia. Murrelet 60:16-22. Suryan, R.M., D.B. Irons, M. Kaufman, J. Benson, P.G.R. Jodice, D.D. Roby, and E.D. Brown. 2002. Short term fluctuations in forage fish availability and the effect on prey selection and broodrearing in the black-legged kittiwake (Rissa tridactyla). Marine Ecology Progress Series 236: 273 287. Wilson, U.W. 1977. A study of the biology of the rhinoceros auklet on Protection Island, Washington. MS thesis, Univ. Washington, Seattle. 8-7

Standardized Productivity Codes: List of Productivity Codes (RHINOCEROS AUKLETS) Always use CAPITAL LETTERS for productivity codes See list of Important Rules to Follow for more details on correct use B Bird Adult bird occupying a nest site, with no egg or chick present. Used when the observer is sure the bird has no egg or chick (this code means a site is confirmed empty other than the adult bird, there is no need to combine B and N codes into BN). For kittiwakes, used between the time a nest is constructed and the first egg is laid, or after egg(s) or chick(s) is lost. BU Bird w/ Unknown Adult bird occupying a site, with no egg or chick visible/felt. Used when the observer cannot see or feel the entire nest contents to be sure whether there is an egg, a chick, or nothing. BE Bird w/ Egg Adult bird with an egg. E Egg Egg present, with no adult. Use numbers and/or + to indicate more than one (e.g., E2+ = at least two eggs) Use standardized modifiers to describe special egg status (e.g., Ed = dead egg) BC Bird w/ Chick Adult bird with chick. C Chick Chick present, with no adult. Use numbers and/or + to indicate more than one (e.g., C2+ = at least two chicks) Use standardized modifiers to describe special egg status (e.g., Cd = dead chick) U Unknown Nest site with nothing clearly visible. Seldom used - only when the observer is not sure of the nest contents (e.g., cliff nest site obscured by fog or other birds, crevice nest site offering a poor, incomplete view, etc.). If an observer records U many times, especially at crucial times (hatch and fledge), the nest site may not be included in analysis. N Nest Empty nest site. Used when an egg or chick that was in the nest has been lost and no adult is present. For kittiwakes, this code indicates that a nest structure from the current year is physically present, either before eggs are laid or after the nest fails. NC Not Checked Used between the previous and current check, when a site was not checked (e.g., it was skipped on purpose) or could not be found on that date. This code does not have to be used at the very beginning or end of the season before checks begin or after checks end for a particular nest. 8-8

Standardized Productivity Codes: List of Modifiers Always use lowercase letters for modifiers! See list of Important Rules to Follow for more details on correct use Modifiers to egg status codes Eo Egg lay observed Observer sees egg being laid; used only when event was actually observed, not simply suspected lay Ep Egg pipped Ed Egg dead Hole in egg, sometimes chick bill poking through; hatch date determined to be following day Egg is obviously damaged or broken Ej Egg ejected Egg once in a nest ejected outside the nest cup Ely Egg last year Egg assumed to be from last year from appearance or other evidence Modifiers to chick status codes Co Chick hatch Observer sees chick hatch; used only when event was actually observed, not for pipping observed eggs, observations of newly-hatched wet chicks, or other reasons leading to suspected probable hatch. Do not combine Co modifier with Cw (see below) if you observe a hatching event that results in a wet chick; simply use Co to indicate hatch and not Cow/Cwo Cw Chick wet Cd Chick dead Newly-hatched wet chick observed; indicates chick hatched that day but hatching event was not actually observed Chick actually observed dead (not simply disappeared from nest) Cf Chick fledged Chick actually observed in the act of flying (or jumping for murres) from the nest; very rare! Additional clues to nest fate sh eggshells Presence of fresh eggshell fragments in the nest. Used ONLY with Bird Unknown or Unknown status codes (BUsh or Ush) to give more information about potential hatch; there is no need to use with known status codes (such as C or N) poop poop Presence of fresh poop in or at the entrance of the nest. Used ONLY with Unknown status code (Upoop) to give more information about potential presence of a chick; there is no need to use with known-status codes (such as C or N) call chick calling Chick heard calling but not actually observed. Used ONLY with Bird Unknown or Unknown status codes (BUcall or Ucall) to give more information about potential presence of a chick; not appropriate to use with known-status codes (such as C or N) 8-9

Standardized Productivity Codes: Important Rules to Follow Use capital letters for basic productivity codes and lowercase letters for modifiers. Do not use superscripts or subscripts for any modifiers. Do not put spaces between any characters. Use numbers to indicate quantities of birds, eggs, or chicks greater than one (do NOT use 1 to indicate single numbers). Numbers should always go AFTER the code that the number describes. e.g., B2E means two adult birds with a single egg BE2 means a single adult bird with two eggs Use a plus sign (+) when you can see at least some but not all nest contents in multiple-egg clutches. As with numbers, the plus sign should always go AFTER the letter code (and when more than one, also after the number) that is being described. e.g., BE+ means an adult bird and at least one egg were observed but entire nest contents could not be seen to determine if there was anything more BE2+ means an adult bird and at least two eggs were observed but entire nest contents could not be seen to determine if there was anything more Some codes and modifiers can be combined when appropriate as long as order (see below) is maintained. For instance, with multiple-egg species, if one egg hatches before another so that there is an adult bird present with both an egg and a check, enter BEC. The specific order in which these codes and modifiers are used is VERY IMPORTANT. The codes BEC and BCE, or BE2+ and BE+2, may mean the same thing to us but they are NOT the same to the computer database. Follow these rules for the correct order of codes and modifiers: (1) Always write codes in the order of adult - egg - chick. e.g., BE and not EB (2) When adding modifiers (numbers, letters, or symbols that give more information), always follow the order of: main productivity code - letter modifier - number - plus sign e.g., BEp2+ for adult bird with at least two pipped eggs BE2Cd for adult bird with two eggs and one dead chick B2E+Cd for two adult birds with at least one egg and one dead chick **NOTE** These standardized codes and modifiers should be used for ALL species, even those that we are still summarizing by hand this year (e.g., storm-petrels, cormorants, oystercatchers, gulls, ancient murrelets). The ultimate goal in future years is to use the database to summarize productivity data for all species and the more consistent we can make the data now, the easier it will be to make that transition. 8-10

14.5 cm ~ 9 cm ¾ ~ 7 cm 9.5 cm Won t fit? Smaller section fits? Whole thing fits? < 9.5cm = Small 9.5 14.5 cm = Medium > 14.5cm = Large Figure 1. Diagram of burrow-measuring device and examples of burrows with entrances of different sizes. 8-11

Figure 2. Example of map of rhinoceros auklet burrow locations within a plot. Circled numbers indicate flag numbers. 8-12

Mass - Weigh birds using a Pesola scale. Weigh the bag and bird and then the bag separately at end. Check bag occasionally for dryness. Always hold by the ring when weighing. Hold scale at eye level in the lee of the wind. DON T hold it here! Wing chord - holding the wing next to the body and using a metal ruler with a stop, measure to the furthest feather tip possible. Always measure the right wing. NOT Wing Chord Flattened Wing Measure from here (primary wing coverts) (primaries) to here Diagonal tarsus - length between the intertarsal joint and the distal end of the last leg scale before the toes emerge. Always measure the right leg. Measure from here NOT to here Figure 3. Diagram of rhinoceros auklet chick measurements taken on the Alaska Maritime National Wildlife Refuge. 8-13

Rhinoceros auklet productivity at Chowiet Island Attachment A. Chowiet Island specifics (includes Table A1) PROCEDURE DETAILS SPECIFIC TO CHOWIET Rhinoceros auklet productivity at Chowiet is estimated using indirect evidence of apparent occupancy on three plots at the South Bay colony and three plots at the Spruce Cove colony. Historically, some reproductive success data were collected at Chowiet by directly monitoring accessible nests using flashlights and/or burrow cameras, but this has been halted due to concerns of erosion and disturbance at the colony. Plots used for reproductive success are also used for population indices and data are collected simultaneously (Table A1). See Burrow-nester Population Protocol for data collection procedures. To calculate overall residency index (f/b), use data from second visit for the number of medium/large burrows with apparent occupancy Early (b) and data from the third visit for the number of medium/large burrows with apparent occupancy Late (f) that also had evidence of apparent occupancy at the early-season check (b). Specific Requirements for Chowiet All data collected at the same time as rhinoceros auklet population monitoring (see Burrow-nester Population Protocol) 8-14

Rhinoceros auklet productivity at Chowiet Island Table A1. Summary of work conducted during three visits to rhinoceros auklet plots for population and reproductive success indices at Chowiet Island. The first and third visits are used for calculating population indices (density and apparent occupancy); the second and third visits are used for estimating reproductive success. Visit: 1 2 3 Dates: Late May to mid-june Late June to early July Mid- to late August Tasks: Flag, measure size of, and count all burrows use for density Check all medium/large burrows for evidence of apparent occupancy use for reproductive success Check all medium/large burrows for evidence of apparent occupancy use for reproductive success and apparent occupancy 8-15

at St. Lazaria Island Attachment B. St. Lazaria Island specifics (includes Figure B1) PROCEDURE DETAILS SPECIFIC TO ST. LAZARIA Rhinoceros auklet productivity at St. Lazaria is estimated using indirect evidence of apparent occupancy. Phenology and chick growth are measured by locating burrows (both within and outside of plots) that contain eggs or chicks, which are then followed through the duration of the breeding season. In recent years, most phenology data at St. Lazaria come from hatch dates calculated from chick growth measurements, as we do not begin monitoring many nests until after chicks hatch. Reproductive success: Reproductive success data come from apparent occupancy surveys on population Plots BN1-1, BN1-2, and BN1-3 (Figure B1) done as part of population monitoring (see Burrow-nester Population Protocol for details). Burrows are surveyed for evidence of apparent occupancy twice during the season: around hatching (late June to mid-july) and during late chick-rearing (late July to mid-august). To calculate overall residency index (f/b), use data from first visit for the number of medium/large burrows with apparent occupancy Early (b) and data from the second visit for the number of medium/large burrows with apparent occupancy Late (f). Chronology and chick growth: On St. Lazaria, chronology and chick growth were historically monitored in population plots (BN1-1, BN1-2, and BN1-3), artificial burrows located in various locations throughout the colony, and random burrow sites (referred to as Rhino Night, Rhino Valley, Rhino Lookout, Rhino Point). NEW IN 2018: In order to reduce disturbance in population plots, use only artificial burrows and random burrow sites for chick growth and chronology monitoring in 2018. This will likely result in lower sample sizes than in previous years. The desired sample size for chronology and chick growth is 30, although this is difficult to attain at St. Lazaria (especially when no longer using population plots for accessible nests). If your sample size is small, you can attempt to increase your sample size by searching for new nests. You can also try digging an excavation in burrows that had strong evidence of apparent occupancy (e.g., nest of Angelica leaves found, lots of guano inside, etc.). Excavations are holes dug above the nest chamber to facilitate viewing or reaching into the nest. While they cause additional disturbance, it is probably the way to access rhinoceros auklet chicks at St. Lazaria. We are attempting to eliminate the use of excavations for other burrow-nester data collection at St. Lazaria due to the disturbance they cause, but it is probably the only way to access rhinoceros auklet chicks. NOTE to help us assess the necessity and effects of excavations, please record which burrows did or did not use excavations in 2018. Only create an excavation in thick, stable soils. To create a new excavation, try to determine the location of the nest chamber by using a stick to extend your reach and establish the direction of the tunnel (go gently to avoid harming any burrow occupants!). Once you have determined the general location, use a pointed trowel to dig down towards where you think the tunnel or chamber may be, being careful remove small chunks of soil at a time. Use a folding saw to cut through roots, as needed. Continue digging until you have broken through the soil to the tunnel or nest chamber. At that point, you should be able to reach in as far as needed to reach into the nest (or as far as possible in all directions to determine where next to dig, if the nest is especially long). After determining nest status and/or extracting a chick for measurement, close all excavations by using excavation covers and hardware cloth (secured with staples), and sprinkle with soil to restore it to a natural looking state. Use an ORANGE flag to mark each excavation. Flags should always be placed on the uphill side of each excavation, and inserted in the soil securely to keep crows, ravens, and rhinos from pulling them up. Some burrows may have excavations already from past years. Before checking a previous excavation, cover the front of all burrow entrances using a wood cover secured with staples (or use a daypack) to prevent adults that may be home from flushing from the burrow. Once you have finished assessing the burrow, be sure to remove blockages from all entrances. Begin checking previously-marked artificial burrows and random burrows in early July (preferably the first 5-7 days in July). At the first visit: 8-16

at St. Lazaria Island Mark burrows with a uniquely numbered YELLOW, RED, or PINK flag placed to the right of the entrance. Use Sharpie permanent markers to write on all flags. If the burrow has more than one entrance and/or an excavation, record that information in your data notebook and, if helpful, write it on the entrance flag as well. Use ORANGE flags to mark excavations above (uphill from) the excavation site, and label them with the same number as their corresponding burrow entrances. Keep in mind that winter weather can destroy or move flags between seasons. It is also helpful to create a map showing the location of burrow entrances, to facilitate finding them later in the season. Determine an initial nest status for each nest by flashlight or grubbing. Historically at St. Lazaria, nest status data were recorded both in a notebook/or PDA and on the nest flag itself with marker; beginning in 2018, you need to record nest status data ONLY in a Rite-in-the-Rain notebook, NOT on the flag. For nests that appear occupied but you cannot directly see/access, try using a historic excavation or digging a new one to access the nest chamber. Be sure to record which nests used excavations. If there is already a chick present (with no attendant adult and allowable weather conditions), remove the chick and take all required measurements. Thereafter, check nests every 5-7 days throughout the end of the season. Determine nest status and measure chicks each time. New in 2018: we are recording relaxed wing chord measurements only, no longer flattened wing chord. Special note: Rhinoceros auklets are very susceptible to human disturbance. While in the colony, voices should be kept to a whisper and stomping avoided. This is imperative while working in the rhinoceros auklet colony to avoid flushing adults from their burrows and to avoid crushing burrows. While walking, it is best to place your feet near the base of tussocks where the roots are most supportive. This is particularly true if there are wide gaps and exposed soil between tussocks. Test any area you suspect may be too fragile to walk on by gingerly pressing with one foot to see if the soil starts to cave in. If the burrow entrance is nearby that leads to this questionable spot, you can reach into it to get an idea of the thickness of the roof of the burrow. You may want to use flags or flagging to mark the appropriate route, or conversely, to avoid certain areas. The combination of guano and wet soil conditions on moderate or even lower angle slopes can make work in the colony difficult. Additionally, if the soils have become heavily saturated from rain the soil can become even more sensitive to your weight. As such, avoid working at the colony during very wet conditions. Crows can be savvy at St. Lazaria and may attempt to dig into nests after watching crews extract and measure chicks. If there are crows around, please consider hiding your actions from them as much as possible; in some years, an umbrella has been used to shield activities from the crows. 8-17

at St. Lazaria Island Specific Requirements for St. Lazaria Dates: Early July to late August/early September (every 5-7 days). Optimal sample size: 30 chicks, however, this is very difficult to obtain, so try for as many as possible; start with burrows excavated in previous years. Time of day: Any time. Weather: Phenology: Any weather, but for your sake, try to avoid excessively wet and windy conditions if possible, given that grubbing requires minimal clothing. Chick growth: Avoid wet and windy conditions. Equipment needed: Phenology: Grubbing gauntlets, raincoat with sleeves cut off, Nitrile gloves with the finger tips cut off to about the first knuckle (optional), plot maps, Rite-in-the-Rain notebook, flags (for initial visit; white for plot boundaries, pink, red, or yellow for burrows, and orange for excavations), two flashlights with extra batteries, two permanent markers, two pencils. Creating excavations: trowel, folding saw, hardware cloth, metal shears, staples, excavation covers, orange flags, Sharpie permanent marker, plot map. Chick growth: Bird bags or pantyhose toes, Pesola scales (60g, 100g, 300 g), wing ruler (150mm), calipers, umbrella, Rite-in-the-Rain notebook, pencils, paper towels or rags (for drying muddy hands before handling chicks). Figure B1. Locations of rhinoceros auklet sites used for productivity, phenology and chick growth at St. Lazaria Island. 8-18

Protocol Revision History Log Protocol Revision History Log Revision Date Changes made New version # Jan 2018 April 2017 Removed measurement of flattened wing (sometimes done at St. Lazaria) and modified St. Lazaria section for clarity and to eliminate chick sampling in population plots Specified that a leap year Julian date calendar should be used in leap years, clarified that for phenology calculations we require confirmed visualization of the empty nest site, egg, or chick less than or equal to 7 days apart for that site to be used Jan 2016 Fixed page number references in text 1.2 April 2014 Changed font to Arial, added revision history log, replaced revision date with version # on first page, added protocol # to first page, changed number format of tables and figures in island attachments, changed page number format to include protocol #, made minor grammatical edits, added clarification to main protocol about recording evidence types and runways, clarified that final calculations should use subset of nests occupied early, added clarification to Chowiet appendix about runways, clarified St. Lazaria sections, and corrected chick growth rate equation. May 2013 Protocol developed in standardized format from historic protocols, includes Chowiet and St.Lazaria attachments 1.0 1.4 1.3 1.1 8-19